CN106651966B - Picture color identification method and system - Google Patents
Picture color identification method and system Download PDFInfo
- Publication number
- CN106651966B CN106651966B CN201610858967.2A CN201610858967A CN106651966B CN 106651966 B CN106651966 B CN 106651966B CN 201610858967 A CN201610858967 A CN 201610858967A CN 106651966 B CN106651966 B CN 106651966B
- Authority
- CN
- China
- Prior art keywords
- color
- pixel points
- value
- channel
- target
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Landscapes
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The invention relates to a picture color identification method and a system, wherein the picture color identification method comprises the following steps: converting the target picture into a set color space, and acquiring channel values of all pixel points of the target picture in all color channels in the color space; respectively calculating membership degrees of the pixel points among a plurality of color clustering centers according to channel values corresponding to the pixel points, and clustering the pixel points of the target picture according to the membership degrees to obtain a plurality of categories of pixel points; and respectively determining the color of the pixel points of each category, and identifying the color of the target picture according to the color of the pixel points. The method and the system for identifying the color of the picture can clearly and accurately identify the color of the blurred picture, and effectively improve the accuracy of the color identification of the picture.
Description
Technical Field
The invention relates to the technical field of image processing, in particular to a method and a system for identifying picture colors.
Background
In video monitoring or other processes requiring picture identification, it is often necessary to identify the color of a target picture (target picture) to identify the corresponding picture.
In a conventional scheme, the target picture is usually converted into a certain color space, and the color of the target picture is identified according to the color channel value corresponding to each pixel point in the color space by obtaining the channel value corresponding to each color channel of each pixel point in the target picture. However, for a blurry picture, the method is used for corresponding color identification, and the defect of low accuracy exists.
Disclosure of Invention
Therefore, it is necessary to provide a method and a system for identifying a color of a picture, aiming at the technical problem that a traditional method is used for identifying a color of a blurry picture and the accuracy is low.
A picture color identification method comprises the following steps:
converting the target picture into a set color space, and acquiring channel values of all pixel points of the target picture in all color channels in the color space;
respectively calculating membership degrees of the pixel points among a plurality of color clustering centers according to channel values corresponding to the pixel points, and clustering the pixel points of the target picture according to the membership degrees to obtain a plurality of categories of pixel points;
and respectively determining the color of the pixel points of each category, and identifying the color of the target picture according to the color of the pixel points.
A picture color recognition system comprising:
the acquisition module is used for converting the target picture into a set color space and acquiring channel values of all pixel points of the target picture in all color channels in the color space;
the clustering module is used for respectively calculating the membership degrees of the pixel points among a plurality of color clustering centers according to the channel values corresponding to the pixel points, and clustering the pixel points of the target picture according to the membership degrees to obtain a plurality of categories of pixel points;
and the identification module is used for respectively determining the color of the pixel points of each category and identifying the color of the target picture according to the color of the pixel points.
According to the image color identification method and the image color identification system, after the target image is converted into the set color space, clustering is carried out according to the channel values of all the pixel points in the color space so as to obtain multiple types of pixel points with extremely high similarity, and then color identification is carried out on the various types of pixel points with higher similarity so as to determine the color of the target image.
Drawings
FIG. 1 is a flowchart of a method for identifying colors of a picture according to an embodiment;
FIG. 2 is a diagram illustrating a BP network training process according to an embodiment;
FIG. 3 is a video surveillance image of an embodiment;
FIG. 4 is a diagram illustrating clustered sub-image pixel points according to an embodiment;
fig. 5 is a schematic structural diagram of a picture color identification system according to an embodiment.
Detailed Description
The following describes in detail specific embodiments of the picture color recognition method and system according to the present invention with reference to the accompanying drawings.
Referring to fig. 1, fig. 1 is a flowchart illustrating a method for detecting a width of a touch pattern according to an embodiment, including the following steps:
s10, converting the target picture into a set color space, and acquiring channel values of each pixel point of the target picture in each color channel in the color space;
the set color space may include a Lab color space. The Lab color space comprises three color channels L, a and b; the L channel is used for expressing the brightness of the pixel, the value range is [0,100], and the L channel expresses from pure black to pure white; the range of the channel a from red to green is [ -128,127 ]; the b channel represents the range from yellow to blue, with values ranging from-128,127.
The target picture is converted into a Lab color space, and the L channel value, the a channel value and the b channel value of each pixel point of the target picture are respectively obtained, so that the brightness information and the color information of the pixel points can be fully obtained.
S20, calculating membership degrees of the pixel points among a plurality of color clustering centers according to channel values corresponding to the pixel points, and clustering the pixel points of the target picture according to the membership degrees to obtain a plurality of categories of pixel points;
the color clustering centers are preset in corresponding color spaces and can be set to be 5 color clustering centers, and the 5 color clustering centers respectively correspond to different color channel values in the corresponding color spaces. The membership degree indicates the degree of the pixel point belonging to the color cluster center. And corresponding membership degrees of certain pixel points relative to all the color clustering centers exist, and after clustering is carried out, the pixel points belong to the category of the color clustering center corresponding to the maximum value of the membership degree. The sum of the membership degrees of each pixel belonging to each color cluster center is 1, that isWherein n is the number of pixel points.
And S30, determining the color of each pixel point of each category respectively, and identifying the color of the target picture according to the color of the pixel points.
The steps can respectively convert the pixel points of each category into HSV color space, and obtain the H channel value, the S channel value and the V channel value of each pixel point in the HSV color space. And substituting the H channel value, the S channel value and the V channel value of the pixel point into a back-propagation (error back-propagation) training BP network to carry out color identification on the pixel point so as to improve the accuracy of the identified color.
According to the image color identification method provided by the embodiment, after the target image is converted into the set color space, clustering is performed according to the channel value corresponding to each pixel point in the color space, so that multiple types of pixel points with extremely high similarity are obtained, and then color identification is performed on the various types of pixel points with higher similarity, so that the color of the target image is determined, clear and accurate color identification can be performed on the fuzzy image, and the accuracy of image color identification is effectively improved.
In an embodiment, before the step of calculating the membership degrees of the pixel points among the plurality of color clustering centers according to the channel values corresponding to the pixel points, and clustering the pixel points of the target picture according to the membership degrees to obtain the pixel points of a plurality of categories, the method may further include:
reading the color types of the target picture, and setting a plurality of color clustering centers according to the color types; wherein each color clustering center has different channel values in the color space.
The color corresponding to the color clustering center can be set as each representative color corresponding to the target picture, and the channel value corresponding to the color clustering center in the corresponding color space is the channel value corresponding to each representative color in the target picture. For example, the three color channels of the Lab color space may be subjected to transform processing such as normalization, the channel values of the color channels are respectively transformed into numerical value intervals [0,255], and the channel values corresponding to the color cluster centers are respectively set to cn1(0, 255), cn2(0,255,0), cn3(100,100,100), cn4(100,0,255), and cn5(100,255,255).
In an embodiment, the step of calculating the membership degrees of the pixel points among the plurality of color clustering centers according to the channel values corresponding to the pixel points, and clustering the pixel points of the target picture according to the membership degrees to obtain the pixel points of a plurality of categories may include:
calculating d between jth pixel point and ith color cluster centerijWherein d isij=||ci-xj||,ciTaking the value of the channel of the ith color cluster center in each channel of the color space, xjTaking the value of the j-th pixel point in each channel of the color space, dijIs the euclidean distance, i.e. between the jth pixel point and the ith color cluster center; c is | |i-xjI stands for ci-xjA modulus value of (d);
according to dijCalculating the membership degree u between the jth pixel point and the ith color cluster centerijWherein, in the step (A),c is the color clustering center number, m is a weighting index, and the weighting index m can be set to any value in a numerical range [1, ∞), namely m ∈ [1, ∞);
according to dijAnd uijJudging whether the target function meets constraint conditions or not;
if so, u isijSubstituting the cluster center into a cluster center updating formula to update the cluster centern is the total number of pixel points of the target picture;
respectively calculating the membership degrees between the pixel points and each clustering center;
and judging the pixel points to be the category corresponding to the color clustering center corresponding to the maximum membership degree.
X is abovejAnd taking values of the channels of the jth pixel point in the corresponding color space, wherein the values comprise three color channel values. For example, if the color space is a Lab color space, x isjCan be (L)x,ax,bx) Similarly, when the color space is a Lab color space, ciCan be (L)c,ac,bc)。
As an example, according to dijAnd uijThe process of determining whether the objective function satisfies the constraint condition may include:
Judging whether the target value is smaller than a first preset threshold value;
and if so, judging that the target function meets the constraint condition.
The first preset threshold may be set according to the weighting index m, for example, to 1 or 1.5.
In one embodiment, the above is according to dijAnd uijAfter the step of judging whether the objective function meets the constraint condition, the method may further include:
a. if the target function does not satisfy the constraint condition, u is addedijSubstituting the cluster center into a cluster center updating formula to update the cluster center
b. Calculating the Euclidean distance d between the ith pixel point and the updated jth color cluster centerij‘;
c. According to dij' calculating membership u between ith pixel point and updated jth color cluster centerij‘;
d. According to dij' and uij' judging whether the target function meets the constraint condition;
e. if not, entering the step a until the target function meets the constraint condition.
In this embodiment, when the target function does not satisfy the constraint condition, that is, when the pixels in the target picture are accurately and sufficiently clustered, the target function is classifiedUpdating the color clustering centers, and determining d from the pixel points to each color clustering center according to the updated color clustering centersij' and uijAnd', until the target function meets the constraint condition, each pixel point is accurately clustered, so that the clustering accuracy of the pixel points is ensured, and the color identification accuracy of the subsequent target picture is ensured.
As an example, according to dijAnd uijThe process of determining whether the objective function satisfies the constraint condition may include:
Calculating the target value and taking the previous group d of the target functionijAnd uijJudging whether the target difference value is smaller than a second preset threshold value or not according to the target difference value between the previous target values;
and if so, judging that the target function meets the constraint condition.
The former group dijAnd uijD before the last clusteringijAnd uij. The second preset threshold may be set according to the weighting index m, for example, to 0.1 or 0.2. And if the target difference value is smaller than a second preset threshold value, indicating that the current clustering reaches the optimal clustering, and performing corresponding color identification according to the clustering result.
In an embodiment, the process of determining the color of each category of pixel point respectively may include:
substituting the channel values corresponding to all pixel points in any type of pixel points into a color identification formula to calculate the color identification value of the pixel point, wherein the color identification formula isxiTaking the value of the channel corresponding to the ith pixel point, WijFor the weighting factor of the BP network input layer to the hidden layer, WjkFor BP networksHidden layer to output layer weight coefficients, functions
Obtaining the y with the highest value frequency from the selected type of pixel pointsiValue according to y having the highest frequencyiAnd determining the color of the pixel points according to the color corresponding to the value.
In this embodiment, the color identification formula is a color identification formula based on a BP network, and the process of training and identifying by substituting part of the pixel points into the BP network may be as shown in fig. 2, with reference to fig. 2, by substituting x into the BP networki(as in x in the figure)1、x2Or x3Etc.) input BP network, and after the input layer, hidden layer and output layer are processed by their corresponding color identification formulas, each x can be outputiCorresponding to yi(e.g. y in the figure)1、y2Or y3Etc.). W is as described aboveijAnd WjkThe reading can be performed at the correspondingly trained BP network. Each y isiThe values correspond to a color channel value in a corresponding color space, and y with the highest value frequencyiThe color channel value corresponding to the value is the color channel value of the pixel of the category.
As an embodiment, before the step of substituting the channel value corresponding to each pixel point in any type of pixel point into the color identification formula to calculate the color identification value of the pixel point, the method may include:
respectively converting all the pixel points into HSV color spaces, and acquiring H component values, S component values and V component values of all the pixel points in the HSV color spaces;
and normalizing the H component value, the S component value and the V component value.
HSV (Hue, Saturation, Value) is a color space created according to the intuitive properties of color, which includes three color channels of Hue (H), Saturation (S) and lightness (V).
In the embodiment, after clustering of each pixel point of the target picture is realized in the Lab color space, each pixel point is converted into the HSV color space to calculate the color identification value, so that the final identification result is processed by integrating the corresponding processing of the two color spaces, and the color accuracy of the identified target picture is further ensured.
In an embodiment, the target picture may be a plurality of sub-pictures obtained by dividing the video surveillance image.
According to the embodiment, the video monitoring image is divided into the plurality of sub-images, each sub-image is used as the target image for color recognition, and on the basis of improving the recognition accuracy, the efficiency of color recognition of the video monitoring image can be improved.
As shown in fig. 3, if the size of the video surveillance image is M × N, the video surveillance image may be equally divided into a × b blocks by rows and columns, where the size of each sub-image is M × N, M is a × M, and N is b × N, and then the color recognition method is respectively used to perform color recognition on a × b sub-images. A schematic diagram of a sub-image after clustering of pixels can be shown in fig. 4.
Referring to fig. 5, fig. 5 is a schematic diagram of a result of a picture color recognition system according to an embodiment, including:
the acquiring module 10 is configured to convert a target picture into a set color space, and acquire a channel value of each pixel point of the target picture in each color channel in the color space;
the clustering module 20 is configured to calculate membership degrees of the pixel points among a plurality of color clustering centers according to channel values corresponding to the pixel points, and cluster the pixel points of the target picture according to the membership degrees to obtain a plurality of categories of pixel points;
and the identification module 30 is configured to determine colors of the pixel points of each category, and identify a color of the target picture according to the colors of the pixel points.
The image color recognition system provided by the invention corresponds to the image color recognition method provided by the invention one by one, and the technical characteristics and the beneficial effects described in the embodiment of the image color recognition method are both suitable for the embodiment of the image color recognition system, so that the statement is made.
The technical features of the embodiments described above may be arbitrarily combined, and for the sake of brevity, all possible combinations of the technical features in the embodiments described above are not described, but should be considered as being within the scope of the present specification as long as there is no contradiction between the combinations of the technical features.
The above-mentioned embodiments only express several embodiments of the present invention, and the description thereof is more specific and detailed, but not construed as limiting the scope of the invention. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the inventive concept, which falls within the scope of the present invention. Therefore, the protection scope of the present patent shall be subject to the appended claims.
Claims (10)
1. A picture color identification method is characterized by comprising the following steps:
converting the target picture into a set color space, and acquiring channel values of all pixel points of the target picture in all color channels in the color space; wherein the color space is a Lab color space; acquiring an L channel value, an a channel value and a b channel value of each pixel point of the target picture;
respectively calculating membership degrees of the pixel points among a plurality of color clustering centers according to channel values corresponding to the pixel points, and clustering the pixel points of the target picture according to the membership degrees to obtain a plurality of categories of pixel points;
respectively determining the color of each category of pixel points, and identifying the color of a target picture according to the color of the pixel points; wherein, the step of determining the color of each category of pixel point comprises: respectively converting the pixel points of each category into an HSV color space, and acquiring an H channel value, an S channel value and a V channel value of each pixel point in the HSV color space; substituting the H channel value, the S channel value and the V channel value of the pixel point into a training BP network for color identification; the process of respectively determining the colors of the pixel points of each category comprises the following steps: all the pixels in any type of pixel pointsSubstituting the corresponding channel values into a color identification formula to calculate the color identification values of the pixel points, wherein the color identification formula isxiTaking the value of the channel corresponding to the ith pixel point, WijFor the weighting factor of the BP network input layer to the hidden layer, WjkFor weighting coefficients, functions, from hidden layer to output layer of BP network
Obtaining the y with the highest value frequency from the selected type of pixel pointsiValue according to y having the highest frequencyiAnd determining the color of the pixel points according to the color corresponding to the value.
2. The method for identifying the color of the picture according to claim 1, wherein before the steps of calculating the membership degrees of the pixel points belonging to a plurality of color clustering centers according to the channel values corresponding to the pixel points, clustering the pixel points of the target picture according to the membership degrees to obtain a plurality of categories of the pixel points, the method further comprises:
reading the color types of the target picture, and setting a plurality of color clustering centers according to the color types; wherein each color clustering center has different channel values in the color space.
3. The method for identifying the color of the picture according to claim 2, wherein the step of calculating the membership degrees of the pixel points among a plurality of color clustering centers according to the channel values corresponding to the pixel points, and clustering the pixel points of the target picture according to the membership degrees to obtain the pixel points of a plurality of categories comprises:
calculating d between jth pixel point and ith color cluster centerijWherein d isij=||ci-xj||,ciCommunication of ith color cluster center in each channel of color spaceValue of trace, xjTaking the value of the j-th pixel point in each channel of the color space, dijIs the Euclidean distance;
according to dijCalculating the membership degree u between the jth pixel point and the ith color cluster centerijWherein, in the step (A),c is the color clustering center number, and m is the weighting index;
according to dijAnd uijJudging whether the target function meets constraint conditions or not;
if so, u isijSubstituting the cluster center into a cluster center updating formula to update the cluster centern is the total number of pixel points of the target picture;
respectively calculating the membership degrees between the pixel points and each clustering center;
and judging the pixel points to be the category corresponding to the color clustering center corresponding to the maximum membership degree.
4. The picture color identification method according to claim 3, wherein the d is a function ofijAnd uijThe process of judging whether the target function meets the constraint condition comprises the following steps:
Judging whether the target value is smaller than a first preset threshold value;
and if so, judging that the target function meets the constraint condition.
5. The picture color identification method according to claim 3, wherein the d is a function ofijAnd uijEyes of the peopleThe step of judging whether the target function meets the constraint condition further comprises the following steps:
a. if the target function does not satisfy the constraint condition, u is addedijSubstituting the cluster center into a cluster center updating formula to update the cluster center
b. Calculating the Euclidean distance d between the ith pixel point and the updated jth color cluster centerij‘;
c. According to dij' calculating membership u between ith pixel point and updated jth color cluster centerij‘;
d. According to dij' and uij' judging whether the target function meets the constraint condition;
e. if not, entering the step a until the target function meets the constraint condition.
6. The picture color identification method according to claim 5, wherein the d is a function ofijAnd uijThe process of judging whether the target function meets the constraint condition comprises the following steps:
Calculating the target value and taking the previous group d of the target functionijAnd uijJudging whether the target difference value is smaller than a second preset threshold value or not according to the target difference value between the previous target values;
and if so, judging that the target function meets the constraint condition.
7. The method for recognizing the color of the picture according to claim 1, wherein the step of substituting the channel values corresponding to the pixel points in any type of the pixel points into the color recognition formula to calculate the color recognition values of the pixel points comprises the steps of:
respectively converting all the pixel points into HSV color spaces, and acquiring H component values, S component values and V component values of all the pixel points in the HSV color spaces;
and normalizing the H component value, the S component value and the V component value.
8. The picture color identification method according to any one of claims 1 to 7, wherein the target picture is a plurality of sub-pictures obtained by dividing a video surveillance image.
9. A picture color recognition system, comprising:
the acquisition module is used for converting the target picture into a set color space and acquiring channel values of all pixel points of the target picture in all color channels in the color space; wherein the color space is a Lab color space; acquiring an L channel value, an a channel value and a b channel value of each pixel point of the target picture;
the clustering module is used for respectively calculating the membership degrees of the pixel points among a plurality of color clustering centers according to the channel values corresponding to the pixel points, and clustering the pixel points of the target picture according to the membership degrees to obtain a plurality of categories of pixel points;
the identification module is used for respectively determining the color of each type of pixel point and identifying the color of the target picture according to the color of the pixel point; wherein, the step of determining the color of each category of pixel point comprises: respectively converting the pixel points of each category into an HSV color space, and acquiring an H channel value, an S channel value and a V channel value of each pixel point in the HSV color space; substituting the H channel value, the S channel value and the V channel value of the pixel point into a training BP network for color identification; and the color identification value calculation module is also used for substituting the channel values corresponding to all the pixel points in any type of pixel points into a color identification formula to calculate the color identification value of the pixel point, wherein the color identification formula isxiTaking the value of the channel corresponding to the ith pixel point, WijFor the weighting factor of the BP network input layer to the hidden layer, WjkFor weighting coefficients, functions, from hidden layer to output layer of BP network
Obtaining the y with the highest value frequency from the selected type of pixel pointsiValue according to y having the highest frequencyiAnd determining the color of the pixel points according to the color corresponding to the value.
10. The system of claim 9, further configured to convert each type of pixel into HSV color space, and obtain an H component value, an S component value, and a V component value of each pixel in HSV color space; and normalizing the H component value, the S component value and the V component value.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610858967.2A CN106651966B (en) | 2016-09-26 | 2016-09-26 | Picture color identification method and system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610858967.2A CN106651966B (en) | 2016-09-26 | 2016-09-26 | Picture color identification method and system |
Publications (2)
Publication Number | Publication Date |
---|---|
CN106651966A CN106651966A (en) | 2017-05-10 |
CN106651966B true CN106651966B (en) | 2020-06-23 |
Family
ID=58854520
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610858967.2A Active CN106651966B (en) | 2016-09-26 | 2016-09-26 | Picture color identification method and system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106651966B (en) |
Families Citing this family (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110349223B (en) * | 2018-04-08 | 2021-04-30 | 中兴通讯股份有限公司 | Image processing method and device |
CN110781904B (en) * | 2019-10-22 | 2022-08-02 | 四川大学 | Vehicle color recognition method and device, computer equipment and readable storage medium |
CN111340897B (en) * | 2020-02-24 | 2023-06-09 | 威斯泰(苏州)智能科技有限公司 | Color identification method based on fuzzy mathematic similarity |
CN115588115A (en) * | 2022-09-27 | 2023-01-10 | 北京羽乐创新科技有限公司 | Method and device for identifying trademark picture |
CN115797494B (en) * | 2023-02-06 | 2023-05-23 | 武汉精臣智慧标识科技有限公司 | Picture binarization processing method and device, electronic equipment and storage medium |
CN116563395B (en) * | 2023-07-12 | 2023-09-15 | 腾讯科技(深圳)有限公司 | Image color feature extraction method and device, storage medium and electronic equipment |
Family Cites Families (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102722720B (en) * | 2012-05-25 | 2014-02-26 | 苏州大学 | Video background extraction method based on hue-saturation-value (HSV) space on-line clustering |
US9299009B1 (en) * | 2013-05-13 | 2016-03-29 | A9.Com, Inc. | Utilizing color descriptors to determine color content of images |
CN103310201B (en) * | 2013-06-26 | 2016-03-23 | 武汉烽火众智数字技术有限责任公司 | The recognition methods of target blend color |
-
2016
- 2016-09-26 CN CN201610858967.2A patent/CN106651966B/en active Active
Also Published As
Publication number | Publication date |
---|---|
CN106651966A (en) | 2017-05-10 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106651966B (en) | Picture color identification method and system | |
CN108108761B (en) | Rapid traffic signal lamp detection method based on deep feature learning | |
US8774503B2 (en) | Method for color feature extraction | |
CN106803257B (en) | Method for segmenting disease spots in crop disease leaf image | |
CN112906649B (en) | Video segmentation method, device, computer device and medium | |
US7528991B2 (en) | Method of generating a mask image of membership of single pixels to certain chromaticity classes and of adaptive improvement of a color image | |
CN111160216B (en) | Living body face recognition method with multiple characteristics and multiple models | |
CN109191428B (en) | Masking texture feature-based full-reference image quality evaluation method | |
CN111178120B (en) | Pest image detection method based on crop identification cascading technology | |
CN107944359A (en) | Flame detecting method based on video | |
CN115457004B (en) | Intelligent detection method of zinc paste based on computer vision | |
CN107292307B (en) | Automatic identification method and system for inverted Chinese character verification code | |
CN107180439B (en) | Color cast characteristic extraction and color cast detection method based on Lab chromaticity space | |
CN112906550B (en) | Static gesture recognition method based on watershed transformation | |
CN114926661B (en) | Textile surface color data processing and identifying method and system | |
JP4901229B2 (en) | Red-eye detection method, apparatus, and program | |
CN113408573B (en) | Method and device for automatically classifying and classifying tile color numbers based on machine learning | |
CN111079637A (en) | Method, device and equipment for segmenting rape flowers in field image and storage medium | |
Shi et al. | The objective evaluation of image object segmentation quality | |
CN110020673A (en) | A kind of method of HSV color space color identification and noise filtering | |
CN113052821B (en) | Quality evaluation method for power equipment inspection picture | |
CN111563536B (en) | Bamboo strip color self-adaptive classification method based on machine learning | |
CN112381761B (en) | Robust low-illumination enhanced image quality evaluation method | |
JP2020047082A (en) | Information processing apparatus, information processing method, and program | |
CN114820707A (en) | Calculation method for camera target automatic tracking |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |